In [1]:
from google.colab import drive
drive.mount('/content/drive')

Mounted at /content/drive
In [2]:
#Importing Libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px

%matplotlib inline
pd.set_option('display.max_columns', 26)
# Removes the limit for the number of displayed columns
pd.set_option("display.max_columns", None)
# Sets the limit for the number of displayed rows
pd.set_option("display.max_rows", 200)
# setting the precision of floating numbers to 5 decimal points
pd.set_option("display.float_format", lambda x: "%.5f" % x)

# split the data into train and test
from sklearn.model_selection import train_test_split

sns.set()

# to build linear regression_model using statsmodels
import statsmodels.api as sm

# to check model performance
from sklearn.metrics import mean_absolute_error, mean_squared_error

#To ignore unecessary errors
import warnings
warnings.filterwarnings("ignore")
#warnings.simplefilter("ignore", ConvergenceWarning)

# To get diferent metric scores
from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.tools.tools import add_constant
from sklearn import metrics
from sklearn.metrics import roc_curve,auc
from sklearn.metrics import precision_recall_curve
from sklearn.metrics import accuracy_score
from sklearn.metrics import precision_score
from sklearn.metrics import roc_auc_score
from sklearn.metrics import make_scorer
from sklearn.metrics import f1_score
from sklearn.metrics import recall_score
from sklearn.metrics import confusion_matrix
from sklearn.ensemble import RandomForestClassifier
from statsmodels.tools.sm_exceptions import ConvergenceWarning
from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.tools.tools import add_constant
from sklearn.tree import DecisionTreeClassifier
from sklearn import tree
from statsmodels.tools.sm_exceptions import ConvergenceWarning
from sklearn.model_selection import cross_val_score
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
from sklearn.metrics import roc_auc_score, roc_curve
from sklearn.model_selection import KFold, StratifiedKFold
from imblearn.over_sampling import SMOTE
from sklearn.metrics import average_precision_score
import sklearn.linear_model
from sklearn.model_selection import GridSearchCV, train_test_split

import pylab as pl
import scipy.optimize as opt
from sklearn import preprocessing
In [3]:
# To access to the dataset
data='/content/drive/MyDrive/Colab Notebooks/bank-full.csv'
In [4]:
# To read the dataset
df=pd.read_csv(data, delimiter=';')
In [5]:
# let us check the first 5 rows of the dataset
df.head()
Out[5]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome y
0 58 management married tertiary no 2143 yes no unknown 5 may 261 1 -1 0 unknown no
1 44 technician single secondary no 29 yes no unknown 5 may 151 1 -1 0 unknown no
2 33 entrepreneur married secondary no 2 yes yes unknown 5 may 76 1 -1 0 unknown no
3 47 blue-collar married unknown no 1506 yes no unknown 5 may 92 1 -1 0 unknown no
4 33 unknown single unknown no 1 no no unknown 5 may 198 1 -1 0 unknown no
In [6]:
# let us check the last 5 rows of the dataset
df.tail()
Out[6]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome y
45206 51 technician married tertiary no 825 no no cellular 17 nov 977 3 -1 0 unknown yes
45207 71 retired divorced primary no 1729 no no cellular 17 nov 456 2 -1 0 unknown yes
45208 72 retired married secondary no 5715 no no cellular 17 nov 1127 5 184 3 success yes
45209 57 blue-collar married secondary no 668 no no telephone 17 nov 508 4 -1 0 unknown no
45210 37 entrepreneur married secondary no 2971 no no cellular 17 nov 361 2 188 11 other no
In [7]:
# Let us check the dataset shape
df.shape
Out[7]:
(45211, 17)
  • The original dataset has 45211 rows and 17 columns
In [8]:
#checking duplicated values
df.duplicated().sum()
Out[8]:
0
  • There is no duplicated values in the dataset
In [9]:
#checking for null values
df.isnull().sum()
Out[9]:
age          0
job          0
marital      0
education    0
default      0
balance      0
housing      0
loan         0
contact      0
day          0
month        0
duration     0
campaign     0
pdays        0
previous     0
poutcome     0
y            0
dtype: int64
  • There is no null value in the raw dataset
In [10]:
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 45211 entries, 0 to 45210
Data columns (total 17 columns):
 #   Column     Non-Null Count  Dtype 
---  ------     --------------  ----- 
 0   age        45211 non-null  int64 
 1   job        45211 non-null  object
 2   marital    45211 non-null  object
 3   education  45211 non-null  object
 4   default    45211 non-null  object
 5   balance    45211 non-null  int64 
 6   housing    45211 non-null  object
 7   loan       45211 non-null  object
 8   contact    45211 non-null  object
 9   day        45211 non-null  int64 
 10  month      45211 non-null  object
 11  duration   45211 non-null  int64 
 12  campaign   45211 non-null  int64 
 13  pdays      45211 non-null  int64 
 14  previous   45211 non-null  int64 
 15  poutcome   45211 non-null  object
 16  y          45211 non-null  object
dtypes: int64(7), object(10)
memory usage: 5.9+ MB
  • The raw dataset has integer, object variables
In [11]:
# Extracting categorical and numerical columns

cat_cols = [col for col in df.columns if df[col].dtype == 'object']
num_cols = [col for col in df.columns if df[col].dtype != 'object']
In [12]:
# looking at unique values in categorical columns

for col in cat_cols:
    print(f"{col} has {df[col].unique()} values\n")

job has ['management' 'technician' 'entrepreneur' 'blue-collar' 'unknown'
 'retired' 'admin.' 'services' 'self-employed' 'unemployed' 'housemaid'
 'student'] values

marital has ['married' 'single' 'divorced'] values

education has ['tertiary' 'secondary' 'unknown' 'primary'] values

default has ['no' 'yes'] values

housing has ['yes' 'no'] values

loan has ['no' 'yes'] values

contact has ['unknown' 'cellular' 'telephone'] values

month has ['may' 'jun' 'jul' 'aug' 'oct' 'nov' 'dec' 'jan' 'feb' 'mar' 'apr' 'sep'] values

poutcome has ['unknown' 'failure' 'other' 'success'] values

y has ['no' 'yes'] values
In [13]:
df.describe().T
Out[13]:
count mean std min 25% 50% 75% max
age 45211.00000 40.93621 10.61876 18.00000 33.00000 39.00000 48.00000 95.00000
balance 45211.00000 1362.27206 3044.76583 -8019.00000 72.00000 448.00000 1428.00000 102127.00000
day 45211.00000 15.80642 8.32248 1.00000 8.00000 16.00000 21.00000 31.00000
duration 45211.00000 258.16308 257.52781 0.00000 103.00000 180.00000 319.00000 4918.00000
campaign 45211.00000 2.76384 3.09802 1.00000 1.00000 2.00000 3.00000 63.00000
pdays 45211.00000 40.19783 100.12875 -1.00000 -1.00000 -1.00000 -1.00000 871.00000
previous 45211.00000 0.58032 2.30344 0.00000 0.00000 0.00000 0.00000 275.00000
  • The mean age of candidate contacted is 41 years old
  • The average yearly balance is 1362.27 euro
  • The average number of contacts performed during this campaign is of 2.7 ( ~ 3 times)
In [14]:
df.columns
Out[14]:
Index(['age', 'job', 'marital', 'education', 'default', 'balance', 'housing',
       'loan', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays',
       'previous', 'poutcome', 'y'],
      dtype='object')

I. Exploratory Data Analysis (EDA)**¶

I.1. Univariate Analysis¶

In [15]:
# defining functions to create plot

def violin(col):
    fig = px.violin(df, y=col, x="class", color="class", box=True, template = 'plotly_dark')
    return fig.show()

def kde(col):
    grid = sns.FacetGrid(df, hue="class", height = 6, aspect=2)
    grid.map(sns.kdeplot, col)
    grid.add_legend()

def scatter(col1, col2):
    fig = px.scatter(df, x=col1, y=col2, color="class", template = 'plotly_dark')
    return fig.show()
In [16]:
def univar_vis(fd):  # Define univariate visualization function
    title = fd.name
    fig, axes = plt.subplots(2, 2, figsize=(10, 6))
    fig.suptitle(title.upper() + "   " + "Distribution")
    sns.distplot(fd, color="green", bins=5, ax=axes[0, 0], hist=None)
    sns.boxplot(fd, ax=axes[0, 1])
    sns.violinplot(fd, ax=axes[1, 0])
    sns.histplot(fd, ax=axes[1, 1])

    axes[0, 0].axvline(fd.mean(), color="black", linewidth=0.7)
    axes[0, 0].axvline(fd.median(), color="red", linewidth=0.3)
    axes[0, 1].axvline(fd.median(), color="red", linewidth=0.9)
    axes[0, 1].axvline(fd.mean(), color="purple", linewidth=0.7)
    axes[1, 0].axvline(fd.mean(), color="purple", linewidth=0.7)
    axes[1, 0].axvline(fd.median(), color="green", linewidth=0.7)
    axes[1, 1].axvline(fd.mean(), color="purple", linewidth=0.7)
    axes[1, 1].axvline(fd.median(), color="green", linewidth=0.7)
    plt.tight_layout()
    plt.show()
In [17]:
univar_vis(df["age"])
In [18]:
univar_vis(df["balance"])
In [19]:
univar_vis(df["day"])
In [20]:
univar_vis(df["duration"])
In [21]:
univar_vis(df["pdays"])
In [22]:
univar_vis(df["duration"])
In [23]:
univar_vis(df["campaign"])
In [24]:
univar_vis(df["previous"])
In [25]:
# checking numerical features distribution

plt.figure(figsize = (20, 15))
plotnumber = 1

for column in num_cols:
    if plotnumber <= 14:
        ax = plt.subplot(3, 5, plotnumber)
        sns.distplot(df[column])
        plt.xlabel(column)

    plotnumber += 1

plt.tight_layout()
plt.show()

Skewness is present in some of the columns.

In [26]:
# looking at categorical columns

plt.figure(figsize = (20, 10))
plotnumber = 1

for column in cat_cols:
    if plotnumber <= 11:
        ax = plt.subplot(3, 4, plotnumber)
        sns.countplot(df[column], palette = 'bright')
        plt.xlabel(column)

    plotnumber += 1

plt.tight_layout()
plt.show()

I.2. Bivariate Analysis¶

In [27]:
# Display columns
df.columns
Out[27]:
Index(['age', 'job', 'marital', 'education', 'default', 'balance', 'housing',
       'loan', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays',
       'previous', 'poutcome', 'y'],
      dtype='object')
In [28]:
# looking at unique values in categorical columns

for col in cat_cols:
    print(f"{col} has {df[col].unique()} values\n")

job has ['management' 'technician' 'entrepreneur' 'blue-collar' 'unknown'
 'retired' 'admin.' 'services' 'self-employed' 'unemployed' 'housemaid'
 'student'] values

marital has ['married' 'single' 'divorced'] values

education has ['tertiary' 'secondary' 'unknown' 'primary'] values

default has ['no' 'yes'] values

housing has ['yes' 'no'] values

loan has ['no' 'yes'] values

contact has ['unknown' 'cellular' 'telephone'] values

month has ['may' 'jun' 'jul' 'aug' 'oct' 'nov' 'dec' 'jan' 'feb' 'mar' 'apr' 'sep'] values

poutcome has ['unknown' 'failure' 'other' 'success'] values

y has ['no' 'yes'] values
In [29]:
sns.countplot(y ='job', data=df, hue='y', palette = 'bright')  #Job vs y

plt.show()
  • High percentage of Blue collar, managers and technicians jobs have not subscribed
  • Given its number of actual managers, they also appears to be more convinced and who have mostly deposited compared to the other jobs.
In [30]:
sns.countplot(y ='marital', data=df, hue='y', palette = 'bright')  #marital vs y

plt.show()

Married are predominant and the majority have not been more convinced, hence have not deposited. However, there have also more deposit compared to single and divorced

In [31]:
sns.countplot(y ='education', data=df, hue='y', palette = 'bright')  #education vs y

plt.show()

Candidates with secondary education level have more deposit, followed by candidates with tertiary then primary

In [32]:
sns.countplot(y ='contact', data=df, hue='y', palette = 'bright')  #contact vs y


plt.show()

Most of candidates convinced were contacted via cellular

In [33]:
sns.countplot(y ='poutcome', data=df, hue='y', palette = 'bright')  # poutcome vs y
plt.show()

poutcome is higher from unknown compared to those who succeeded, followed by those who failed

In [34]:
fig,axes = plt.subplots(5,2,figsize=(12,15))
for idx,cat_cols in enumerate(num_cols):
    row,col = idx//2,idx%2
    sns.boxplot(y=cat_cols,data=df,x='y',ax=axes[row,col], palette = 'bright')


plt.subplots_adjust(hspace=1)
In [35]:
# pair plot showing relationsh amoung variables
sns.pairplot(data=df, hue='y', corner=True)
from time import sleep
from tqdm import tqdm
for i in tqdm (range (10)):
  sleep(3)

100%|██████████| 10/10 [00:30<00:00,  3.01s/it]
In [36]:
#heatmap of data

plt.figure(figsize=(15,10))

sns.heatmap(df.corr(), annot=True, linewidths=2, linecolor='lightgrey')

plt.show()

Balance and duration have high correlation compared to the others

In [37]:
ax = sns.catplot(
    y="job", x="balance", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="JOB TYPE vs. BALANCE");

The average yearly salary of candidates in management and unknown category is equally high compared to the others, even though we observed an exception with retired people who have highest balance.

In [38]:
ax = sns.catplot(
    y="education", x="balance", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="EDUCATION vs. BALANCE");

Candidates with tertiary education earn more compared to the others

In [39]:
ax = sns.catplot(
    y="marital", x="balance", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="MARITAL vs. BALANCE");

Married candidates have highest average yearly balance compared to single and divorced candidates

In [40]:
ax = sns.catplot(
    y="job", x="duration", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="JOB vs. DURATION");
In [41]:
ax = sns.catplot(
    y="marital", x="duration", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="EDUCATION vs. DURATION");

Singles followed by divorced were the most being in contact during the marketing campaign

In [42]:
ax = sns.catplot(
    y="marital", x="campaign", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="MARITAL vs. CAMPAIGN");
In [43]:
ax = sns.catplot(
    y="job", x="campaign", data=df, kind="bar", height=4.5, aspect=3, palette = 'bright'
)

ax.set_xticklabels(rotation=0).set(
    title="JOB vs. CAMPAIGN");

II. DATA PRE PROCESSING¶

II.1. Data cleaning and transformation¶

In [44]:
Defaults = df.copy()
In [45]:
print(Defaults["campaign"].value_counts())
print(Defaults["job"].value_counts())

1     17544
2     12505
3      5521
4      3522
5      1764
6      1291
7       735
8       540
9       327
10      266
11      201
12      155
13      133
14       93
15       84
16       79
17       69
18       51
19       44
20       43
21       35
22       23
25       22
23       22
24       20
29       16
28       16
26       13
31       12
27       10
32        9
30        8
33        6
34        5
36        4
35        4
43        3
38        3
37        2
50        2
41        2
46        1
58        1
55        1
63        1
51        1
39        1
44        1
Name: campaign, dtype: int64
blue-collar      9732
management       9458
technician       7597
admin.           5171
services         4154
retired          2264
self-employed    1579
entrepreneur     1487
unemployed       1303
housemaid        1240
student           938
unknown           288
Name: job, dtype: int64

II.2. Prepare dataset for modeling¶

In [46]:
df.dtypes
Out[46]:
age           int64
job          object
marital      object
education    object
default      object
balance       int64
housing      object
loan         object
contact      object
day           int64
month        object
duration      int64
campaign      int64
pdays         int64
previous      int64
poutcome     object
y            object
dtype: object
In [47]:
from sklearn.impute import SimpleImputer

# Identify columns with missing values
columns_with_missing = df.columns[df.isnull().any()]

# Impute numerical columns with mean
numerical_cols = df.select_dtypes(include='number').columns
df[numerical_cols] = df[numerical_cols].fillna(df[numerical_cols].median())

# Impute categorical columns with mode
categorical_cols = df.select_dtypes(include='object').columns
df[categorical_cols] = df[categorical_cols].fillna(df[categorical_cols].mode().iloc[0])

# Verify if there are still any missing values
remaining_missing = df.isnull().sum().sum()
print(f"Remaining missing values: {remaining_missing}")

Remaining missing values: 0
In [48]:
df.head()
Out[48]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome y
0 58 management married tertiary no 2143 yes no unknown 5 may 261 1 -1 0 unknown no
1 44 technician single secondary no 29 yes no unknown 5 may 151 1 -1 0 unknown no
2 33 entrepreneur married secondary no 2 yes yes unknown 5 may 76 1 -1 0 unknown no
3 47 blue-collar married unknown no 1506 yes no unknown 5 may 92 1 -1 0 unknown no
4 33 unknown single unknown no 1 no no unknown 5 may 198 1 -1 0 unknown no
In [49]:
df = df.rename(columns={'y': 'deposit'})

# Or rename the existing DataFrame (rather than creating a copy)
#df.rename(columns={'y': 'newName1', 'oldName2': 'newName2'}, inplace=True)
In [50]:
df.head()
Out[50]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome deposit
0 58 management married tertiary no 2143 yes no unknown 5 may 261 1 -1 0 unknown no
1 44 technician single secondary no 29 yes no unknown 5 may 151 1 -1 0 unknown no
2 33 entrepreneur married secondary no 2 yes yes unknown 5 may 76 1 -1 0 unknown no
3 47 blue-collar married unknown no 1506 yes no unknown 5 may 92 1 -1 0 unknown no
4 33 unknown single unknown no 1 no no unknown 5 may 198 1 -1 0 unknown no
In [51]:
# replace the ‘deposit' column contains
# the values 'yes' and 'no'  with
# 1 and  0:
df['deposit'] = df['deposit'].map(
                   {'yes':1 ,'no':0})
In [52]:
df.head()
Out[52]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome deposit
0 58 management married tertiary no 2143 yes no unknown 5 may 261 1 -1 0 unknown 0
1 44 technician single secondary no 29 yes no unknown 5 may 151 1 -1 0 unknown 0
2 33 entrepreneur married secondary no 2 yes yes unknown 5 may 76 1 -1 0 unknown 0
3 47 blue-collar married unknown no 1506 yes no unknown 5 may 92 1 -1 0 unknown 0
4 33 unknown single unknown no 1 no no unknown 5 may 198 1 -1 0 unknown 0
In [53]:
#Let's identify the categorical features
df.columns[df.dtypes == object]
Out[53]:
Index(['job', 'marital', 'education', 'default', 'housing', 'loan', 'contact',
       'month', 'poutcome'],
      dtype='object')
In [54]:
# apply get_dummies function
df_encoded = pd.get_dummies(df, columns=['job', 'marital', 'education', 'default', 'housing', 'loan', 'contact',
       'month', 'poutcome'])
df_encoded.head()
Out[54]:
age balance day duration campaign pdays previous deposit job_admin. job_blue-collar job_entrepreneur job_housemaid job_management job_retired job_self-employed job_services job_student job_technician job_unemployed job_unknown marital_divorced marital_married marital_single education_primary education_secondary education_tertiary education_unknown default_no default_yes housing_no housing_yes loan_no loan_yes contact_cellular contact_telephone contact_unknown month_apr month_aug month_dec month_feb month_jan month_jul month_jun month_mar month_may month_nov month_oct month_sep poutcome_failure poutcome_other poutcome_success poutcome_unknown
0 58 2143 5 261 1 -1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1
1 44 29 5 151 1 -1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 1 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1
2 33 2 5 76 1 -1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 1 0 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1
3 47 1506 5 92 1 -1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1
4 33 1 5 198 1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1
In [55]:
y = df_encoded.deposit
X = df_encoded.drop("deposit", axis = 1)

III. MODELS¶

III.1. Decision Tree¶

  • We want to predict if the client will subscribe a term deposit (variable y).
  • Before we proceed to build a model, we'll have to encode categorical features.
  • We'll split the data into train and test to be able to evaluate the model that we build on the train data.

III.1.1. Train-Test Split for decision tree model¶

In [56]:
# Splitting the data into train and test sets in 70:30 ratio
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.30, random_state=1, shuffle = True )
In [57]:
X_train.shape, X_test.shape
Out[57]:
((31647, 51), (13564, 51))
In [58]:
print("Shape of Training set : ", X_train.shape)
print("Shape of test set : ", X_test.shape)
print("Percentage of classes in training set:")
print(y_train.value_counts(normalize=True))
print("Percentage of classes in test set:")
print(y_test.value_counts(normalize=True))

Shape of Training set :  (31647, 51)
Shape of test set :  (13564, 51)
Percentage of classes in training set:
0   0.88188
1   0.11812
Name: deposit, dtype: float64
Percentage of classes in test set:
0   0.88565
1   0.11435
Name: deposit, dtype: float64

III.1.2. Building Decision Tree Model¶

In [59]:
model = DecisionTreeClassifier(random_state=1)
model.fit(X_train, y_train)
Out[59]:
DecisionTreeClassifier(random_state=1)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
DecisionTreeClassifier(random_state=1)
In [60]:
# defining a function to compute different metrics to check performance of a classification model built using sklearn
def model_performance_classification_sklearn(model, predictors, target):
    """
    Function to compute different metrics to check classification model performance

    model: classifier
    predictors: independent variables
    target: dependent variable
    """

    # predicting using the independent variables
    pred = model.predict(predictors)

    acc = accuracy_score(target, pred)  # to compute Accuracy
    recall = recall_score(target, pred)  # to compute Recall
    precision = precision_score(target, pred)  # to compute Precision
    f1 = f1_score(target, pred)  # to compute F1-score

    # creating a dataframe of metrics
    df_perf = pd.DataFrame(
        {"Accuracy": acc, "Recall": recall, "Precision": precision, "F1": f1,},
        index=[0],
    )

    return df_perf
In [61]:
def confusion_matrix_sklearn(model, predictors, target):
    """
    To plot the confusion_matrix with percentages

    model: classifier
    predictors: independent variables
    target: dependent variable
    """
    y_pred = model.predict(predictors)
    cm = confusion_matrix(target, y_pred)
    labels = np.asarray(
        [
            ["{0:0.0f}".format(item) + "\n{0:.2%}".format(item / cm.flatten().sum())]
            for item in cm.flatten()
        ]
    ).reshape(2, 2)

    plt.figure(figsize=(6, 4))
    sns.heatmap(cm, annot=labels, fmt="")
    plt.ylabel("True label")
    plt.xlabel("Predicted label")

III.1.3. Decision Tree sklearn¶

III.1.3.1. Checking Decision Tree model performance on training set¶

In [62]:
confusion_matrix_sklearn(model, X_train, y_train)
In [63]:
decision_tree_perf_train = model_performance_classification_sklearn(
    model, X_train, y_train
)
decision_tree_perf_train
Out[63]:
Accuracy Recall Precision F1
0 1.00000 1.00000 1.00000 1.00000

The model is not performing well on trained dataset

III.1.3.2. Checking Decision Tree model performance on test set¶

In [64]:
confusion_matrix_sklearn(model, X_test, y_test)
In [65]:
decision_tree_perf_test = model_performance_classification_sklearn(
    model, X_test, y_test
)
decision_tree_perf_test
Out[65]:
Accuracy Recall Precision F1
0 0.87651 0.49323 0.46252 0.47738

Accuracy is high when the model is trained on test dataset

Before pruning the tree let's check the important features

Plotting the feature importance of each variable

In [66]:
feature_names = list(X_train.columns)
importances = model.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(8, 10))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="blue", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

Duration and average yearly balance are the most important features in this case of decision tree using sklearn, without pruning it

III.1.4. Pruning decision tree model¶

III.1.4.1. Pre-pruning¶

In [67]:
# Choose the type of classifier.
estimator = DecisionTreeClassifier(random_state=1, class_weight="balanced")

# Grid of parameters to choose from
parameters = {
    "max_depth": np.arange(2, 7, 2),
    "max_leaf_nodes": [50, 75, 150, 250],
    "min_samples_split": [10, 30, 50, 70],
}

# Type of scoring used to compare parameter combinations
acc_scorer = make_scorer(f1_score)

# Run the grid search
grid_obj = GridSearchCV(estimator, parameters, scoring=acc_scorer, cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
estimator = grid_obj.best_estimator_

# Fit the best algorithm to the data.
estimator.fit(X_train, y_train)
Out[67]:
DecisionTreeClassifier(class_weight='balanced', max_depth=6, max_leaf_nodes=50,
                       min_samples_split=50, random_state=1)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
DecisionTreeClassifier(class_weight='balanced', max_depth=6, max_leaf_nodes=50,
                       min_samples_split=50, random_state=1)
In [68]:
decision_tree_tune_perf_train = model_performance_classification_sklearn(
    estimator, X_train, y_train
)
decision_tree_tune_perf_train
Out[68]:
Accuracy Recall Precision F1
0 0.82513 0.82772 0.38753 0.52790

III.1.4.2. Checking performance on the training set¶

In [69]:
confusion_matrix_sklearn(estimator, X_train, y_train)
In [70]:
decision_tree_tune_perf_train = model_performance_classification_sklearn(
    estimator, X_train, y_train
)
decision_tree_tune_perf_train
Out[70]:
Accuracy Recall Precision F1
0 0.82513 0.82772 0.38753 0.52790

We still have a good accuracy when the model sees the train dataset

III.1.4.3. Checking performance on the test set¶

In [71]:
confusion_matrix_sklearn(estimator, X_test, y_test)
In [72]:
decision_tree_tune_perf_test = model_performance_classification_sklearn(
    estimator, X_test, y_test
)
decision_tree_tune_perf_test
Out[72]:
Accuracy Recall Precision F1
0 0.82159 0.81560 0.37217 0.51111

Accuracy remains almost the same when the model sees the test dataset

III.1.4.4. Visualization of the Decision Tree¶

In [73]:
plt.figure(figsize=(20, 10))
out = tree.plot_tree(
    estimator,
    feature_names=feature_names,
    filled=True,
    fontsize=9,
    node_ids=False,
    class_names=None,
)
# below code will add arrows to the decision tree split if they are missing
for o in out:
    arrow = o.arrow_patch
    if arrow is not None:
        arrow.set_edgecolor("black")
        arrow.set_linewidth(1)
plt.show()
In [74]:
# Text report showing the rules of a decision tree -
print(tree.export_text(estimator, feature_names=feature_names, show_weights=True))

|--- duration <= 205.50
|   |--- poutcome_success <= 0.50
|   |   |--- month_mar <= 0.50
|   |   |   |--- month_oct <= 0.50
|   |   |   |   |--- duration <= 124.50
|   |   |   |   |   |--- month_apr <= 0.50
|   |   |   |   |   |   |--- weights: [5320.42, 309.02] class: 0
|   |   |   |   |   |--- month_apr >  0.50
|   |   |   |   |   |   |--- weights: [271.58, 105.83] class: 0
|   |   |   |   |--- duration >  124.50
|   |   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |   |--- weights: [2297.92, 245.52] class: 0
|   |   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |   |--- weights: [1556.33, 863.56] class: 0
|   |   |   |--- month_oct >  0.50
|   |   |   |   |--- duration <= 95.50
|   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |--- weights: [40.25, 0.00] class: 0
|   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |--- weights: [5.10, 8.47] class: 1
|   |   |   |   |--- duration >  95.50
|   |   |   |   |   |--- day <= 20.50
|   |   |   |   |   |   |--- weights: [39.12, 59.26] class: 1
|   |   |   |   |   |--- day >  20.50
|   |   |   |   |   |   |--- weights: [15.31, 165.09] class: 1
|   |   |--- month_mar >  0.50
|   |   |   |--- duration <= 79.00
|   |   |   |   |--- weights: [16.44, 4.23] class: 0
|   |   |   |--- duration >  79.00
|   |   |   |   |--- campaign <= 7.50
|   |   |   |   |   |--- previous <= 4.50
|   |   |   |   |   |   |--- weights: [38.55, 215.89] class: 1
|   |   |   |   |   |--- previous >  4.50
|   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |--- campaign >  7.50
|   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |--- poutcome_success >  0.50
|   |   |--- duration <= 132.50
|   |   |   |--- campaign <= 3.50
|   |   |   |   |--- pdays <= 98.50
|   |   |   |   |   |--- duration <= 80.00
|   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |--- duration >  80.00
|   |   |   |   |   |   |--- weights: [13.61, 80.43] class: 1
|   |   |   |   |--- pdays >  98.50
|   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |--- weights: [30.62, 50.80] class: 1
|   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |--- weights: [9.64, 0.00] class: 0
|   |   |   |--- campaign >  3.50
|   |   |   |   |--- weights: [13.61, 0.00] class: 0
|   |   |--- duration >  132.50
|   |   |   |--- job_entrepreneur <= 0.50
|   |   |   |   |--- pdays <= 202.50
|   |   |   |   |   |--- balance <= -56.50
|   |   |   |   |   |   |--- weights: [0.57, 0.00] class: 0
|   |   |   |   |   |--- balance >  -56.50
|   |   |   |   |   |   |--- weights: [34.59, 524.91] class: 1
|   |   |   |   |--- pdays >  202.50
|   |   |   |   |   |--- weights: [9.64, 42.33] class: 1
|   |   |   |--- job_entrepreneur >  0.50
|   |   |   |   |--- weights: [1.13, 0.00] class: 0
|--- duration >  205.50
|   |--- duration <= 473.50
|   |   |--- contact_unknown <= 0.50
|   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |--- pdays <= 374.50
|   |   |   |   |   |   |--- weights: [1636.84, 922.83] class: 0
|   |   |   |   |   |--- pdays >  374.50
|   |   |   |   |   |   |--- weights: [4.54, 101.60] class: 1
|   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |--- loan_yes <= 0.50
|   |   |   |   |   |   |--- weights: [1200.84, 2933.57] class: 1
|   |   |   |   |   |--- loan_yes >  0.50
|   |   |   |   |   |   |--- weights: [252.30, 126.99] class: 0
|   |   |   |--- poutcome_success >  0.50
|   |   |   |   |--- pdays <= 29.50
|   |   |   |   |   |--- weights: [3.40, 4.23] class: 1
|   |   |   |   |--- pdays >  29.50
|   |   |   |   |   |--- housing_yes <= 0.50
|   |   |   |   |   |   |--- weights: [39.12, 1164.12] class: 1
|   |   |   |   |   |--- housing_yes >  0.50
|   |   |   |   |   |   |--- weights: [29.48, 355.58] class: 1
|   |   |--- contact_unknown >  0.50
|   |   |   |--- month_oct <= 0.50
|   |   |   |   |--- duration <= 368.50
|   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |--- weights: [1207.07, 29.63] class: 0
|   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |--- weights: [3.97, 12.70] class: 1
|   |   |   |   |--- duration >  368.50
|   |   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |   |--- weights: [115.66, 63.50] class: 0
|   |   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |   |--- weights: [222.82, 29.63] class: 0
|   |   |   |--- month_oct >  0.50
|   |   |   |   |--- weights: [0.00, 25.40] class: 1
|   |--- duration >  473.50
|   |   |--- duration <= 647.50
|   |   |   |--- contact_unknown <= 0.50
|   |   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |   |--- month_jul <= 0.50
|   |   |   |   |   |   |--- weights: [374.77, 1515.47] class: 1
|   |   |   |   |   |--- month_jul >  0.50
|   |   |   |   |   |   |--- weights: [138.91, 300.55] class: 1
|   |   |   |   |--- poutcome_success >  0.50
|   |   |   |   |   |--- balance <= 13362.50
|   |   |   |   |   |   |--- weights: [11.34, 304.79] class: 1
|   |   |   |   |   |--- balance >  13362.50
|   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |--- contact_unknown >  0.50
|   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |--- balance <= 2941.00
|   |   |   |   |   |   |--- weights: [60.10, 152.39] class: 1
|   |   |   |   |   |--- balance >  2941.00
|   |   |   |   |   |   |--- weights: [16.44, 8.47] class: 0
|   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |--- duration <= 505.50
|   |   |   |   |   |   |--- weights: [48.19, 16.93] class: 0
|   |   |   |   |   |--- duration >  505.50
|   |   |   |   |   |   |--- weights: [111.13, 114.29] class: 1
|   |   |--- duration >  647.50
|   |   |   |--- contact_unknown <= 0.50
|   |   |   |   |--- duration <= 847.50
|   |   |   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |   |   |--- weights: [213.75, 1439.27] class: 1
|   |   |   |   |   |--- poutcome_success >  0.50
|   |   |   |   |   |   |--- weights: [4.54, 173.56] class: 1
|   |   |   |   |--- duration >  847.50
|   |   |   |   |   |--- duration <= 3193.00
|   |   |   |   |   |   |--- weights: [179.73, 2290.13] class: 1
|   |   |   |   |   |--- duration >  3193.00
|   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |--- contact_unknown >  0.50
|   |   |   |   |--- duration <= 827.50
|   |   |   |   |   |--- duration <= 815.50
|   |   |   |   |   |   |--- weights: [119.06, 330.19] class: 1
|   |   |   |   |   |--- duration >  815.50
|   |   |   |   |   |   |--- weights: [7.94, 0.00] class: 0
|   |   |   |   |--- duration >  827.50
|   |   |   |   |   |--- campaign <= 11.50
|   |   |   |   |   |   |--- weights: [91.28, 732.33] class: 1
|   |   |   |   |   |--- campaign >  11.50
|   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0

Plotting the feature importance of each variable

In [75]:
# importance of features in the tree building

importances = estimator.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(8, 8))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="blue", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

After pre pruning, duration remains the most important features, however in this case, poutcome is the next important feature.

III.1.4.5. Cost complexity pruning (Post pruning)¶

In [76]:
clf = DecisionTreeClassifier(random_state=1, class_weight="balanced")
path = clf.cost_complexity_pruning_path(X_train, y_train)
ccp_alphas, impurities = abs(path.ccp_alphas), path.impurities
In [77]:
pd.DataFrame(path)
Out[77]:
ccp_alphas impurities
0 0.00000 -0.00000
1 0.00000 -0.00000
2 0.00000 -0.00000
3 0.00000 -0.00000
4 0.00000 -0.00000
... ... ...
1583 0.00914 0.30836
1584 0.01229 0.32065
1585 0.02263 0.34329
1586 0.02630 0.39588
1587 0.10412 0.50000

1588 rows × 2 columns

In [78]:
fig, ax = plt.subplots(figsize=(10, 5))
ax.plot(ccp_alphas[:-1], impurities[:-1], marker="o", drawstyle="steps-post")
ax.set_xlabel("effective alpha")
ax.set_ylabel("total impurity of leaves")
ax.set_title("Total Impurity vs effective alpha for training set")
plt.show()

Next, we train a decision tree using effective alphas. The last value in ccp_alphas is the alpha value that prunes the whole tree, leaving the tree, clfs[-1], with one node

In [79]:
clfs = []
for ccp_alpha in ccp_alphas:
    clf = DecisionTreeClassifier(
        random_state=1, ccp_alpha=ccp_alpha, class_weight="balanced"
    )
    clf.fit(X_train, y_train)
    clfs.append(clf)
print(
    "Number of nodes in the last tree is: {} with ccp_alpha: {}".format(
        clfs[-1].tree_.node_count, ccp_alphas[-1]
    )
)

Number of nodes in the last tree is: 1 with ccp_alpha: 0.10412091138731988
In [80]:
clfs = clfs[:-1]
ccp_alphas = ccp_alphas[:-1]

node_counts = [clf.tree_.node_count for clf in clfs]
depth = [clf.tree_.max_depth for clf in clfs]
fig, ax = plt.subplots(2, 1, figsize=(10, 7))
ax[0].plot(ccp_alphas, node_counts, marker="o", drawstyle="steps-post")
ax[0].set_xlabel("alpha")
ax[0].set_ylabel("number of nodes")
ax[0].set_title("Number of nodes vs alpha")
ax[1].plot(ccp_alphas, depth, marker="o", drawstyle="steps-post")
ax[1].set_xlabel("alpha")
ax[1].set_ylabel("depth of tree")
ax[1].set_title("Depth vs alpha")
fig.tight_layout()
III.1.4.5.1. F1 Score vs alpha for training and testing sets¶
In [81]:
f1_train = []
for clf in clfs:
    pred_train = clf.predict(X_train)
    values_train = f1_score(y_train, pred_train)
    f1_train.append(values_train)

f1_test = []
for clf in clfs:
    pred_test = clf.predict(X_test)
    values_test = f1_score(y_test, pred_test)
    f1_test.append(values_test)
In [82]:
fig, ax = plt.subplots(figsize=(15, 5))
ax.set_xlabel("alpha")
ax.set_ylabel("F1 Score")
ax.set_title("F1 Score vs alpha for training and testing sets")
ax.plot(ccp_alphas, f1_train, marker="o", label="train", drawstyle="steps-post")
ax.plot(ccp_alphas, f1_test, marker="o", label="test", drawstyle="steps-post")
ax.legend()
plt.show()
In [83]:
index_best_model = np.argmax(f1_test)
best_model = clfs[index_best_model]
print(best_model)

DecisionTreeClassifier(ccp_alpha=7.258901846204399e-05, class_weight='balanced',
                       random_state=1)
III.1.4.5.2. Checking performance on the training set¶
In [84]:
confusion_matrix_sklearn(best_model, X_train, y_train)
In [85]:
decision_tree_post_perf_train = model_performance_classification_sklearn(
    best_model, X_train, y_train
)
decision_tree_post_perf_train
Out[85]:
Accuracy Recall Precision F1
0 0.93127 0.99866 0.63239 0.77440

Accuracy has slightly improved in post prune case

III.1.4.5.3. Checking performance on the test set¶
In [86]:
confusion_matrix_sklearn(best_model, X_test, y_test)
In [87]:
decision_tree_post_test = model_performance_classification_sklearn(
    best_model, X_test, y_test
)
decision_tree_post_test
Out[87]:
Accuracy Recall Precision F1
0 0.86597 0.73050 0.44730 0.55485

Observations

After post pruning the decision tree the performance has generalized on training and test set. We are getting reasonable accuracy value compared to the other metrics that highly fluctuate.

In [88]:
plt.figure(figsize=(20, 10))

out = tree.plot_tree(
    best_model,
    feature_names=feature_names,
    filled=True,
    fontsize=9,
    node_ids=False,
    class_names=None,
)
for o in out:
    arrow = o.arrow_patch
    if arrow is not None:
        arrow.set_edgecolor("black")
        arrow.set_linewidth(1)
plt.show()
In [89]:
# Text report showing the rules of a decision tree -

print(tree.export_text(best_model, feature_names=feature_names, show_weights=True))

|--- duration <= 205.50
|   |--- poutcome_success <= 0.50
|   |   |--- month_mar <= 0.50
|   |   |   |--- month_oct <= 0.50
|   |   |   |   |--- duration <= 124.50
|   |   |   |   |   |--- month_apr <= 0.50
|   |   |   |   |   |   |--- month_feb <= 0.50
|   |   |   |   |   |   |   |--- month_sep <= 0.50
|   |   |   |   |   |   |   |   |--- age <= 24.50
|   |   |   |   |   |   |   |   |   |--- duration <= 99.50
|   |   |   |   |   |   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [45.92, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |--- duration >  99.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 242.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- balance >  242.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [10.21, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- age >  24.50
|   |   |   |   |   |   |   |   |   |--- day <= 4.50
|   |   |   |   |   |   |   |   |   |   |--- contact_cellular <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [159.88, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- contact_cellular >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- day >  4.50
|   |   |   |   |   |   |   |   |   |   |--- pdays <= 392.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |   |--- pdays >  392.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |--- month_sep >  0.50
|   |   |   |   |   |   |   |   |--- duration <= 90.50
|   |   |   |   |   |   |   |   |   |--- weights: [19.28, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  90.50
|   |   |   |   |   |   |   |   |   |--- day <= 10.50
|   |   |   |   |   |   |   |   |   |   |--- job_services <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [14.74, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- job_services >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |--- day >  10.50
|   |   |   |   |   |   |   |   |   |   |--- day <= 12.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 16.93] class: 1
|   |   |   |   |   |   |   |   |   |   |--- day >  12.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |--- month_feb >  0.50
|   |   |   |   |   |   |   |--- day <= 9.50
|   |   |   |   |   |   |   |   |--- education_tertiary <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [211.48, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- education_tertiary >  0.50
|   |   |   |   |   |   |   |   |   |--- balance <= 388.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [45.92, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  388.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 29.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- age >  29.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |--- day >  9.50
|   |   |   |   |   |   |   |   |--- duration <= 74.50
|   |   |   |   |   |   |   |   |   |--- pdays <= 296.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [15.88, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- pdays >  296.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- duration >  74.50
|   |   |   |   |   |   |   |   |   |--- contact_cellular <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- contact_cellular >  0.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 121.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- duration >  121.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |--- month_apr >  0.50
|   |   |   |   |   |   |--- housing_yes <= 0.50
|   |   |   |   |   |   |   |--- day <= 27.50
|   |   |   |   |   |   |   |   |--- duration <= 96.50
|   |   |   |   |   |   |   |   |   |--- balance <= 294.50
|   |   |   |   |   |   |   |   |   |   |--- day <= 14.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- day >  14.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- balance >  294.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [21.54, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  96.50
|   |   |   |   |   |   |   |   |   |--- age <= 35.00
|   |   |   |   |   |   |   |   |   |   |--- age <= 20.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- age >  20.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 55.03] class: 1
|   |   |   |   |   |   |   |   |   |--- age >  35.00
|   |   |   |   |   |   |   |   |   |   |--- contact_unknown <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- contact_unknown >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |--- day >  27.50
|   |   |   |   |   |   |   |   |--- weights: [20.98, 0.00] class: 0
|   |   |   |   |   |   |--- housing_yes >  0.50
|   |   |   |   |   |   |   |--- day <= 23.50
|   |   |   |   |   |   |   |   |--- duration <= 104.50
|   |   |   |   |   |   |   |   |   |--- job_self-employed <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [152.51, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- job_self-employed >  0.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 129.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |   |--- balance >  129.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.10, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  104.50
|   |   |   |   |   |   |   |   |   |--- age <= 32.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 770.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- balance >  770.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- age >  32.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [27.21, 0.00] class: 0
|   |   |   |   |   |   |   |--- day >  23.50
|   |   |   |   |   |   |   |   |--- day <= 27.50
|   |   |   |   |   |   |   |   |   |--- balance <= 169.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  169.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 12.70] class: 1
|   |   |   |   |   |   |   |   |--- day >  27.50
|   |   |   |   |   |   |   |   |   |--- weights: [8.50, 0.00] class: 0
|   |   |   |   |--- duration >  124.50
|   |   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |   |--- month_sep <= 0.50
|   |   |   |   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |   |   |   |--- pdays <= 18.50
|   |   |   |   |   |   |   |   |   |--- age <= 60.50
|   |   |   |   |   |   |   |   |   |   |--- month_feb <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- month_feb >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- age >  60.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |   |   |   |   |--- pdays >  18.50
|   |   |   |   |   |   |   |   |   |--- pdays <= 93.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 25.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- balance >  25.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- pdays >  93.50
|   |   |   |   |   |   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |   |   |   |--- pdays <= 381.50
|   |   |   |   |   |   |   |   |   |--- balance <= 7102.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- balance >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [930.39, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  7102.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 32.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- age >  32.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [21.54, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- pdays >  381.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |--- month_sep >  0.50
|   |   |   |   |   |   |   |--- day <= 5.00
|   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |--- day >  5.00
|   |   |   |   |   |   |   |   |--- pdays <= 185.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 25.40] class: 1
|   |   |   |   |   |   |   |   |--- pdays >  185.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |   |--- month_apr <= 0.50
|   |   |   |   |   |   |   |--- pdays <= 41.50
|   |   |   |   |   |   |   |   |--- age <= 61.50
|   |   |   |   |   |   |   |   |   |--- age <= 28.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 10.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [17.58, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- balance >  10.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |--- age >  28.50
|   |   |   |   |   |   |   |   |   |   |--- day <= 3.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- day >  3.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |--- age >  61.50
|   |   |   |   |   |   |   |   |   |--- day <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 147.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  147.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [9.64, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- day >  11.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 199.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  199.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |--- pdays >  41.50
|   |   |   |   |   |   |   |   |--- pdays <= 126.50
|   |   |   |   |   |   |   |   |   |--- day <= 16.50
|   |   |   |   |   |   |   |   |   |   |--- job_student <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- job_student >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- day >  16.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 1770.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- balance >  1770.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [11.91, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- pdays >  126.50
|   |   |   |   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 21.17] class: 1
|   |   |   |   |   |   |--- month_apr >  0.50
|   |   |   |   |   |   |   |--- day <= 28.50
|   |   |   |   |   |   |   |   |--- campaign <= 5.50
|   |   |   |   |   |   |   |   |   |--- contact_telephone <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- job_self-employed <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- job_self-employed >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- contact_telephone >  0.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 2808.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- balance >  2808.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 8.47] class: 1
|   |   |   |   |   |   |   |   |--- campaign >  5.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |--- day >  28.50
|   |   |   |   |   |   |   |   |--- duration <= 174.00
|   |   |   |   |   |   |   |   |   |--- weights: [10.21, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  174.00
|   |   |   |   |   |   |   |   |   |--- campaign <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 8.47] class: 1
|   |   |   |   |   |   |   |   |   |--- campaign >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |--- month_oct >  0.50
|   |   |   |   |--- duration <= 95.50
|   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |--- weights: [40.25, 0.00] class: 0
|   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |--- education_tertiary <= 0.50
|   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |--- education_tertiary >  0.50
|   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |--- duration >  95.50
|   |   |   |   |   |--- day <= 20.50
|   |   |   |   |   |   |--- education_primary <= 0.50
|   |   |   |   |   |   |   |--- campaign <= 2.50
|   |   |   |   |   |   |   |   |--- previous <= 5.50
|   |   |   |   |   |   |   |   |   |--- duration <= 130.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 46.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [8.50, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- age >  46.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- duration >  130.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [10.21, 46.56] class: 1
|   |   |   |   |   |   |   |   |--- previous >  5.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |--- campaign >  2.50
|   |   |   |   |   |   |   |   |--- weights: [5.10, 0.00] class: 0
|   |   |   |   |   |   |--- education_primary >  0.50
|   |   |   |   |   |   |   |--- weights: [7.37, 0.00] class: 0
|   |   |   |   |   |--- day >  20.50
|   |   |   |   |   |   |--- campaign <= 3.50
|   |   |   |   |   |   |   |--- balance <= 10745.00
|   |   |   |   |   |   |   |   |--- campaign <= 1.50
|   |   |   |   |   |   |   |   |   |--- weights: [5.67, 139.69] class: 1
|   |   |   |   |   |   |   |   |--- campaign >  1.50
|   |   |   |   |   |   |   |   |   |--- age <= 65.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 41.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- age >  41.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- age >  65.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 16.93] class: 1
|   |   |   |   |   |   |   |--- balance >  10745.00
|   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |--- campaign >  3.50
|   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |--- month_mar >  0.50
|   |   |   |--- duration <= 79.00
|   |   |   |   |--- job_management <= 0.50
|   |   |   |   |   |--- weights: [13.04, 0.00] class: 0
|   |   |   |   |--- job_management >  0.50
|   |   |   |   |   |--- marital_married <= 0.50
|   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |--- marital_married >  0.50
|   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |--- duration >  79.00
|   |   |   |   |--- previous <= 4.50
|   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |--- campaign <= 7.50
|   |   |   |   |   |   |   |--- education_unknown <= 0.50
|   |   |   |   |   |   |   |   |--- duration <= 173.50
|   |   |   |   |   |   |   |   |   |--- duration <= 164.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 31.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- age >  31.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- duration >  164.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  173.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.27, 63.50] class: 1
|   |   |   |   |   |   |   |--- education_unknown >  0.50
|   |   |   |   |   |   |   |   |--- day <= 9.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- day >  9.00
|   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |--- campaign >  7.50
|   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |--- previous >  4.50
|   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |--- poutcome_success >  0.50
|   |   |--- duration <= 132.50
|   |   |   |--- campaign <= 3.50
|   |   |   |   |--- pdays <= 98.50
|   |   |   |   |   |--- duration <= 80.00
|   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |--- duration >  80.00
|   |   |   |   |   |   |--- day <= 27.50
|   |   |   |   |   |   |   |--- month_oct <= 0.50
|   |   |   |   |   |   |   |   |--- pdays <= 84.50
|   |   |   |   |   |   |   |   |   |--- pdays <= 67.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- pdays >  67.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- pdays >  84.50
|   |   |   |   |   |   |   |   |   |--- weights: [6.24, 67.73] class: 1
|   |   |   |   |   |   |   |--- month_oct >  0.50
|   |   |   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |   |   |   |--- day >  27.50
|   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |--- pdays >  98.50
|   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |   |   |   |--- balance <= 849.50
|   |   |   |   |   |   |   |   |   |--- weights: [5.10, 38.10] class: 1
|   |   |   |   |   |   |   |   |--- balance >  849.50
|   |   |   |   |   |   |   |   |   |--- balance <= 3325.00
|   |   |   |   |   |   |   |   |   |   |--- month_feb <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [9.07, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- month_feb >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |--- balance >  3325.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |--- weights: [8.50, 0.00] class: 0
|   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |--- weights: [9.64, 0.00] class: 0
|   |   |   |--- campaign >  3.50
|   |   |   |   |--- weights: [13.61, 0.00] class: 0
|   |   |--- duration >  132.50
|   |   |   |--- job_entrepreneur <= 0.50
|   |   |   |   |--- pdays <= 202.50
|   |   |   |   |   |--- balance <= -56.50
|   |   |   |   |   |   |--- weights: [0.57, 0.00] class: 0
|   |   |   |   |   |--- balance >  -56.50
|   |   |   |   |   |   |--- duration <= 162.50
|   |   |   |   |   |   |   |--- duration <= 158.50
|   |   |   |   |   |   |   |   |--- weights: [14.74, 169.33] class: 1
|   |   |   |   |   |   |   |--- duration >  158.50
|   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.47] class: 1
|   |   |   |   |   |   |--- duration >  162.50
|   |   |   |   |   |   |   |--- weights: [15.31, 347.12] class: 1
|   |   |   |   |--- pdays >  202.50
|   |   |   |   |   |--- pdays <= 277.00
|   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |--- pdays >  277.00
|   |   |   |   |   |   |--- balance <= 3378.00
|   |   |   |   |   |   |   |--- weights: [3.40, 42.33] class: 1
|   |   |   |   |   |   |--- balance >  3378.00
|   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |--- job_entrepreneur >  0.50
|   |   |   |   |--- weights: [1.13, 0.00] class: 0
|--- duration >  205.50
|   |--- duration <= 473.50
|   |   |--- contact_unknown <= 0.50
|   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |--- pdays <= 374.50
|   |   |   |   |   |   |--- month_jun <= 0.50
|   |   |   |   |   |   |   |--- month_mar <= 0.50
|   |   |   |   |   |   |   |   |--- duration <= 389.50
|   |   |   |   |   |   |   |   |   |--- month_sep <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- month_oct <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 19
|   |   |   |   |   |   |   |   |   |   |--- month_oct >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- month_sep >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 25.40] class: 1
|   |   |   |   |   |   |   |   |--- duration >  389.50
|   |   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- month_jan <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- month_jan >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [9.64, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 622.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- balance >  622.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- month_mar >  0.50
|   |   |   |   |   |   |   |   |--- weights: [3.40, 59.26] class: 1
|   |   |   |   |   |   |--- month_jun >  0.50
|   |   |   |   |   |   |   |--- weights: [10.77, 84.66] class: 1
|   |   |   |   |   |--- pdays >  374.50
|   |   |   |   |   |   |--- weights: [4.54, 101.60] class: 1
|   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |--- loan_yes <= 0.50
|   |   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |   |--- balance <= 50.50
|   |   |   |   |   |   |   |   |--- balance <= -46.00
|   |   |   |   |   |   |   |   |   |--- weights: [32.32, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- balance >  -46.00
|   |   |   |   |   |   |   |   |   |--- month_jul <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- month_feb <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- month_feb >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- month_jul >  0.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 410.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  410.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |--- balance >  50.50
|   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |--- month_jan <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |--- month_jan >  0.50
|   |   |   |   |   |   |   |   |   |   |--- day <= 25.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 71.96] class: 1
|   |   |   |   |   |   |   |   |   |   |--- day >  25.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |--- month_sep <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- pdays <= 142.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |   |--- pdays >  142.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [15.31, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- month_sep >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 25.40] class: 1
|   |   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |   |--- campaign <= 1.50
|   |   |   |   |   |   |   |   |--- job_technician <= 0.50
|   |   |   |   |   |   |   |   |   |--- job_management <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- job_management >  0.50
|   |   |   |   |   |   |   |   |   |   |--- education_tertiary <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- education_tertiary >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- job_technician >  0.50
|   |   |   |   |   |   |   |   |   |--- balance <= 9642.50
|   |   |   |   |   |   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [13.61, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- balance >  9642.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.47] class: 1
|   |   |   |   |   |   |   |--- campaign >  1.50
|   |   |   |   |   |   |   |   |--- duration <= 383.50
|   |   |   |   |   |   |   |   |   |--- poutcome_unknown <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- previous <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 25.40] class: 1
|   |   |   |   |   |   |   |   |   |   |--- previous >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- poutcome_unknown >  0.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 59.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- age >  59.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |--- duration >  383.50
|   |   |   |   |   |   |   |   |   |--- pdays <= 43.50
|   |   |   |   |   |   |   |   |   |   |--- marital_single <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- marital_single >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- pdays >  43.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 16.93] class: 1
|   |   |   |   |   |--- loan_yes >  0.50
|   |   |   |   |   |   |--- month_jul <= 0.50
|   |   |   |   |   |   |   |--- day <= 1.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 16.93] class: 1
|   |   |   |   |   |   |   |--- day >  1.50
|   |   |   |   |   |   |   |   |--- day <= 20.50
|   |   |   |   |   |   |   |   |   |--- balance <= 360.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [35.72, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  360.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 2094.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- balance >  2094.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [9.64, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- day >  20.50
|   |   |   |   |   |   |   |   |   |--- month_jan <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- campaign <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- campaign >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- month_jan >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [10.21, 0.00] class: 0
|   |   |   |   |   |   |--- month_jul >  0.50
|   |   |   |   |   |   |   |--- day <= 3.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 8.47] class: 1
|   |   |   |   |   |   |   |--- day >  3.00
|   |   |   |   |   |   |   |   |--- duration <= 425.00
|   |   |   |   |   |   |   |   |   |--- age <= 31.50
|   |   |   |   |   |   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |--- age >  31.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [102.62, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  425.00
|   |   |   |   |   |   |   |   |   |--- duration <= 431.50
|   |   |   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- duration >  431.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [11.34, 0.00] class: 0
|   |   |   |--- poutcome_success >  0.50
|   |   |   |   |--- pdays <= 29.50
|   |   |   |   |   |--- age <= 59.00
|   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |--- age >  59.00
|   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |--- pdays >  29.50
|   |   |   |   |   |--- housing_yes <= 0.50
|   |   |   |   |   |   |--- weights: [39.12, 1164.12] class: 1
|   |   |   |   |   |--- housing_yes >  0.50
|   |   |   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |   |   |--- duration <= 225.50
|   |   |   |   |   |   |   |   |--- day <= 15.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 16.93] class: 1
|   |   |   |   |   |   |   |   |--- day >  15.00
|   |   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |   |   |   |--- duration >  225.50
|   |   |   |   |   |   |   |   |--- weights: [11.34, 283.62] class: 1
|   |   |   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |   |   |--- pdays <= 352.00
|   |   |   |   |   |   |   |   |--- age <= 38.50
|   |   |   |   |   |   |   |   |   |--- balance <= 544.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- balance >  544.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- age >  38.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 42.33] class: 1
|   |   |   |   |   |   |   |--- pdays >  352.00
|   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |--- contact_unknown >  0.50
|   |   |   |--- month_oct <= 0.50
|   |   |   |   |--- duration <= 368.50
|   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |--- pdays <= 37.00
|   |   |   |   |   |   |   |--- month_feb <= 0.50
|   |   |   |   |   |   |   |   |--- age <= 26.50
|   |   |   |   |   |   |   |   |   |--- day <= 29.50
|   |   |   |   |   |   |   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [48.19, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- day >  29.50
|   |   |   |   |   |   |   |   |   |   |--- education_tertiary <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- education_tertiary >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- age >  26.50
|   |   |   |   |   |   |   |   |   |--- day <= 7.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 208.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  208.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- day >  7.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [856.12, 0.00] class: 0
|   |   |   |   |   |   |   |--- month_feb >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |--- pdays >  37.00
|   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |--- day <= 15.00
|   |   |   |   |   |   |   |--- weights: [0.00, 12.70] class: 1
|   |   |   |   |   |   |--- day >  15.00
|   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |   |   |--- duration >  368.50
|   |   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |   |--- balance <= -356.50
|   |   |   |   |   |   |   |--- campaign <= 3.00
|   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |--- campaign >  3.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 12.70] class: 1
|   |   |   |   |   |   |--- balance >  -356.50
|   |   |   |   |   |   |   |--- age <= 38.50
|   |   |   |   |   |   |   |   |--- balance <= 290.00
|   |   |   |   |   |   |   |   |   |--- weights: [17.01, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- balance >  290.00
|   |   |   |   |   |   |   |   |   |--- balance <= 3027.00
|   |   |   |   |   |   |   |   |   |   |--- day <= 4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- day >  4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- balance >  3027.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [7.37, 0.00] class: 0
|   |   |   |   |   |   |   |--- age >  38.50
|   |   |   |   |   |   |   |   |--- balance <= 16.00
|   |   |   |   |   |   |   |   |   |--- age <= 49.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [8.50, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- age >  49.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 426.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 8.47] class: 1
|   |   |   |   |   |   |   |   |   |   |--- duration >  426.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- balance >  16.00
|   |   |   |   |   |   |   |   |   |--- weights: [52.16, 0.00] class: 0
|   |   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |   |--- age <= 30.50
|   |   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |   |--- job_entrepreneur <= 0.50
|   |   |   |   |   |   |   |   |   |--- default_yes <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [40.82, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- default_yes >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- job_entrepreneur >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |   |--- campaign <= 1.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- campaign >  1.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |   |   |--- age >  30.50
|   |   |   |   |   |   |   |--- duration <= 430.50
|   |   |   |   |   |   |   |   |--- weights: [125.30, 0.00] class: 0
|   |   |   |   |   |   |   |--- duration >  430.50
|   |   |   |   |   |   |   |   |--- duration <= 431.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- duration >  431.50
|   |   |   |   |   |   |   |   |   |--- job_blue-collar <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [36.85, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- job_blue-collar >  0.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 3977.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- balance >  3977.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |--- month_oct >  0.50
|   |   |   |   |--- weights: [0.00, 25.40] class: 1
|   |--- duration >  473.50
|   |   |--- duration <= 647.50
|   |   |   |--- contact_unknown <= 0.50
|   |   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |   |--- month_jul <= 0.50
|   |   |   |   |   |   |--- month_jan <= 0.50
|   |   |   |   |   |   |   |--- housing_yes <= 0.50
|   |   |   |   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |   |   |   |--- balance <= 12579.50
|   |   |   |   |   |   |   |   |   |   |--- month_aug <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- month_aug >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- balance >  12579.50
|   |   |   |   |   |   |   |   |   |   |--- contact_telephone <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- contact_telephone >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |   |   |   |--- duration <= 619.00
|   |   |   |   |   |   |   |   |   |   |--- day <= 20.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- day >  20.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- duration >  619.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |   |--- housing_yes >  0.50
|   |   |   |   |   |   |   |   |--- month_apr <= 0.50
|   |   |   |   |   |   |   |   |   |--- campaign <= 5.50
|   |   |   |   |   |   |   |   |   |   |--- job_unemployed <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- job_unemployed >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- campaign >  5.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- month_apr >  0.50
|   |   |   |   |   |   |   |   |   |--- day <= 24.50
|   |   |   |   |   |   |   |   |   |   |--- day <= 18.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- day >  18.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [11.91, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- day >  24.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 29.63] class: 1
|   |   |   |   |   |   |--- month_jan >  0.50
|   |   |   |   |   |   |   |--- day <= 14.50
|   |   |   |   |   |   |   |   |--- weights: [0.57, 12.70] class: 1
|   |   |   |   |   |   |   |--- day >  14.50
|   |   |   |   |   |   |   |   |--- duration <= 482.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 8.47] class: 1
|   |   |   |   |   |   |   |   |--- duration >  482.00
|   |   |   |   |   |   |   |   |   |--- age <= 30.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 530.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- duration >  530.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- age >  30.50
|   |   |   |   |   |   |   |   |   |   |--- job_services <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [23.81, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- job_services >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 4.23] class: 1
|   |   |   |   |   |--- month_jul >  0.50
|   |   |   |   |   |   |--- campaign <= 8.50
|   |   |   |   |   |   |   |--- duration <= 530.50
|   |   |   |   |   |   |   |   |--- duration <= 496.50
|   |   |   |   |   |   |   |   |   |--- loan_yes <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- loan_yes >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  496.50
|   |   |   |   |   |   |   |   |   |--- age <= 41.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 29.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- age >  29.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- age >  41.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [15.88, 0.00] class: 0
|   |   |   |   |   |   |   |--- duration >  530.50
|   |   |   |   |   |   |   |   |--- job_housemaid <= 0.50
|   |   |   |   |   |   |   |   |   |--- balance <= 1596.00
|   |   |   |   |   |   |   |   |   |   |--- balance <= 1112.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- balance >  1112.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  1596.00
|   |   |   |   |   |   |   |   |   |   |--- duration <= 540.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- duration >  540.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [12.47, 76.20] class: 1
|   |   |   |   |   |   |   |   |--- job_housemaid >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |--- campaign >  8.50
|   |   |   |   |   |   |   |--- weights: [6.80, 0.00] class: 0
|   |   |   |   |--- poutcome_success >  0.50
|   |   |   |   |   |--- balance <= 13362.50
|   |   |   |   |   |   |--- month_dec <= 0.50
|   |   |   |   |   |   |   |--- education_primary <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [7.37, 292.09] class: 1
|   |   |   |   |   |   |   |--- education_primary >  0.50
|   |   |   |   |   |   |   |   |--- age <= 55.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- age >  55.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 12.70] class: 1
|   |   |   |   |   |   |--- month_dec >  0.50
|   |   |   |   |   |   |   |--- weights: [0.57, 0.00] class: 0
|   |   |   |   |   |--- balance >  13362.50
|   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |--- contact_unknown >  0.50
|   |   |   |   |--- month_may <= 0.50
|   |   |   |   |   |--- balance <= 2941.00
|   |   |   |   |   |   |--- job_entrepreneur <= 0.50
|   |   |   |   |   |   |   |--- day <= 3.50
|   |   |   |   |   |   |   |   |--- campaign <= 1.50
|   |   |   |   |   |   |   |   |   |--- weights: [5.67, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- campaign >  1.50
|   |   |   |   |   |   |   |   |   |--- duration <= 588.00
|   |   |   |   |   |   |   |   |   |   |--- duration <= 520.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- duration >  520.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- duration >  588.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |--- day >  3.50
|   |   |   |   |   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |   |   |   |   |--- balance <= -514.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  -514.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 595.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  595.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 55.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- age >  55.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |--- job_entrepreneur >  0.50
|   |   |   |   |   |   |   |--- weights: [5.10, 0.00] class: 0
|   |   |   |   |   |--- balance >  2941.00
|   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |--- weights: [15.88, 0.00] class: 0
|   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |--- weights: [0.57, 8.47] class: 1
|   |   |   |   |--- month_may >  0.50
|   |   |   |   |   |--- duration <= 505.50
|   |   |   |   |   |   |--- education_secondary <= 0.50
|   |   |   |   |   |   |   |--- balance <= 154.50
|   |   |   |   |   |   |   |   |--- job_management <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.70, 12.70] class: 1
|   |   |   |   |   |   |   |   |--- job_management >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |--- balance >  154.50
|   |   |   |   |   |   |   |   |--- age <= 52.50
|   |   |   |   |   |   |   |   |   |--- weights: [14.74, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- age >  52.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.13, 4.23] class: 1
|   |   |   |   |   |   |--- education_secondary >  0.50
|   |   |   |   |   |   |   |--- weights: [28.92, 0.00] class: 0
|   |   |   |   |   |--- duration >  505.50
|   |   |   |   |   |   |--- job_management <= 0.50
|   |   |   |   |   |   |   |--- balance <= 279.00
|   |   |   |   |   |   |   |   |--- job_services <= 0.50
|   |   |   |   |   |   |   |   |   |--- job_retired <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- loan_yes <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [40.82, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- loan_yes >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- job_retired >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 4.23] class: 1
|   |   |   |   |   |   |   |   |--- job_services >  0.50
|   |   |   |   |   |   |   |   |   |--- marital_married <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 25.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- age >  25.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 12.70] class: 1
|   |   |   |   |   |   |   |   |   |--- marital_married >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |--- balance >  279.00
|   |   |   |   |   |   |   |   |--- balance <= 630.50
|   |   |   |   |   |   |   |   |   |--- balance <= 549.00
|   |   |   |   |   |   |   |   |   |   |--- balance <= 346.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [6.24, 16.93] class: 1
|   |   |   |   |   |   |   |   |   |   |--- balance >  346.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [10.77, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- balance >  549.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 21.17] class: 1
|   |   |   |   |   |   |   |   |--- balance >  630.50
|   |   |   |   |   |   |   |   |   |--- age <= 26.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |--- age >  26.00
|   |   |   |   |   |   |   |   |   |   |--- duration <= 589.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- duration >  589.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [10.77, 0.00] class: 0
|   |   |   |   |   |   |--- job_management >  0.50
|   |   |   |   |   |   |   |--- age <= 45.50
|   |   |   |   |   |   |   |   |--- campaign <= 3.50
|   |   |   |   |   |   |   |   |   |--- balance <= 831.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 31.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.47] class: 1
|   |   |   |   |   |   |   |   |   |   |--- age >  31.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- balance >  831.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- campaign >  3.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 12.70] class: 1
|   |   |   |   |   |   |   |--- age >  45.50
|   |   |   |   |   |   |   |   |--- weights: [3.97, 0.00] class: 0
|   |   |--- duration >  647.50
|   |   |   |--- contact_unknown <= 0.50
|   |   |   |   |--- duration <= 847.50
|   |   |   |   |   |--- poutcome_success <= 0.50
|   |   |   |   |   |   |--- month_apr <= 0.50
|   |   |   |   |   |   |   |--- balance <= -962.50
|   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |--- balance >  -962.50
|   |   |   |   |   |   |   |   |--- month_nov <= 0.50
|   |   |   |   |   |   |   |   |   |--- balance <= 653.00
|   |   |   |   |   |   |   |   |   |   |--- contact_cellular <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- contact_cellular >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 17
|   |   |   |   |   |   |   |   |   |--- balance >  653.00
|   |   |   |   |   |   |   |   |   |   |--- job_unemployed <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [49.33, 584.17] class: 1
|   |   |   |   |   |   |   |   |   |   |--- job_unemployed >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- month_nov >  0.50
|   |   |   |   |   |   |   |   |   |--- pdays <= 161.50
|   |   |   |   |   |   |   |   |   |   |--- contact_telephone <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [24.95, 131.23] class: 1
|   |   |   |   |   |   |   |   |   |   |--- contact_telephone >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- pdays >  161.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |--- month_apr >  0.50
|   |   |   |   |   |   |   |--- day <= 20.50
|   |   |   |   |   |   |   |   |--- duration <= 666.50
|   |   |   |   |   |   |   |   |   |--- weights: [5.10, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  666.50
|   |   |   |   |   |   |   |   |   |--- duration <= 722.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 1178.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 33.87] class: 1
|   |   |   |   |   |   |   |   |   |   |--- balance >  1178.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- duration >  722.50
|   |   |   |   |   |   |   |   |   |   |--- housing_no <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- housing_no >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 12.70] class: 1
|   |   |   |   |   |   |   |--- day >  20.50
|   |   |   |   |   |   |   |   |--- weights: [1.13, 55.03] class: 1
|   |   |   |   |   |--- poutcome_success >  0.50
|   |   |   |   |   |   |--- weights: [4.54, 173.56] class: 1
|   |   |   |   |--- duration >  847.50
|   |   |   |   |   |--- weights: [180.86, 2290.13] class: 1
|   |   |   |--- contact_unknown >  0.50
|   |   |   |   |--- duration <= 827.50
|   |   |   |   |   |--- duration <= 815.50
|   |   |   |   |   |   |--- age <= 32.50
|   |   |   |   |   |   |   |--- loan_yes <= 0.50
|   |   |   |   |   |   |   |   |--- balance <= 81.00
|   |   |   |   |   |   |   |   |   |--- balance <= -0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 21.17] class: 1
|   |   |   |   |   |   |   |   |   |--- balance >  -0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- balance >  81.00
|   |   |   |   |   |   |   |   |   |--- duration <= 701.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 31.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 25.40] class: 1
|   |   |   |   |   |   |   |   |   |   |--- age >  31.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- duration >  701.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.40, 84.66] class: 1
|   |   |   |   |   |   |   |--- loan_yes >  0.50
|   |   |   |   |   |   |   |   |--- job_admin. <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [6.24, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- job_admin. >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |--- age >  32.50
|   |   |   |   |   |   |   |--- job_technician <= 0.50
|   |   |   |   |   |   |   |   |--- duration <= 687.00
|   |   |   |   |   |   |   |   |   |--- marital_married <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 651.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.23] class: 1
|   |   |   |   |   |   |   |   |   |   |--- duration >  651.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [7.94, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- marital_married >  0.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 486.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- balance >  486.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- duration >  687.00
|   |   |   |   |   |   |   |   |   |--- marital_married <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 731.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- duration >  731.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- marital_married >  0.50
|   |   |   |   |   |   |   |   |   |   |--- age <= 44.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- age >  44.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |--- job_technician >  0.50
|   |   |   |   |   |   |   |   |--- day <= 5.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.40, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- day >  5.50
|   |   |   |   |   |   |   |   |   |--- education_secondary <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- education_secondary >  0.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 747.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- duration >  747.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.57, 33.87] class: 1
|   |   |   |   |   |--- duration >  815.50
|   |   |   |   |   |   |--- weights: [7.94, 0.00] class: 0
|   |   |   |   |--- duration >  827.50
|   |   |   |   |   |--- campaign <= 11.50
|   |   |   |   |   |   |--- marital_divorced <= 0.50
|   |   |   |   |   |   |   |--- balance <= 790.50
|   |   |   |   |   |   |   |   |--- duration <= 834.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- duration >  834.50
|   |   |   |   |   |   |   |   |   |--- age <= 30.50
|   |   |   |   |   |   |   |   |   |   |--- job_technician <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 97.36] class: 1
|   |   |   |   |   |   |   |   |   |   |--- job_technician >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.70, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- age >  30.50
|   |   |   |   |   |   |   |   |   |   |--- duration <= 1870.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- duration >  1870.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.54, 0.00] class: 0
|   |   |   |   |   |   |   |--- balance >  790.50
|   |   |   |   |   |   |   |   |--- age <= 26.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.27, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- age >  26.50
|   |   |   |   |   |   |   |   |   |--- day <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.13, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- day >  2.50
|   |   |   |   |   |   |   |   |   |   |--- balance <= 6327.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [20.98, 279.39] class: 1
|   |   |   |   |   |   |   |   |   |   |--- balance >  6327.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |--- marital_divorced >  0.50
|   |   |   |   |   |   |   |--- weights: [6.24, 135.46] class: 1
|   |   |   |   |   |--- campaign >  11.50
|   |   |   |   |   |   |--- weights: [2.83, 0.00] class: 0

Plotting the feature importance of each variable

In [90]:
importances = best_model.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12, 12))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="Blue", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

Observations from tree

The tree is quite complex as compared to the pre-pruned tree. The feature importance is fairly the same.

III.2. Comparing Decision Tree Models¶

In [91]:
# training performance comparison

models_train_comp_df = pd.concat(
    [
        decision_tree_perf_train.T,
        decision_tree_tune_perf_train.T,
        decision_tree_post_perf_train.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
]
print("Training performance comparison:")
models_train_comp_df

Training performance comparison:
Out[91]:
Decision Tree sklearn Decision Tree (Pre-Pruning) Decision Tree (Post-Pruning)
Accuracy 1.00000 0.82513 0.93127
Recall 1.00000 0.82772 0.99866
Precision 1.00000 0.38753 0.63239
F1 1.00000 0.52790 0.77440
In [92]:
# testing performance comparison

models_test_comp_df = pd.concat(
    [
        decision_tree_perf_test.T,
        decision_tree_tune_perf_test.T,
        decision_tree_post_test.T,
    ],
    axis=1,
)
models_test_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
]
print("Test set performance comparison:")
models_test_comp_df

Test set performance comparison:
Out[92]:
Decision Tree sklearn Decision Tree (Pre-Pruning) Decision Tree (Post-Pruning)
Accuracy 0.87651 0.82159 0.86597
Recall 0.49323 0.81560 0.73050
Precision 0.46252 0.37217 0.44730
F1 0.47738 0.51111 0.55485

Observations

Decision tree model with default parameters is overfitting the training data and is not able to generalize well. Pre-pruned tree has given a generalized performance with balanced values of precision and recall. Post-pruned tree is giving a high F1 score as compared to other models but the difference between precision and recall is high. The bank will be able to maintain a balance between resources, brand equity, candidates who will deposit using the sklearn decision tree model.

VI. Random forest¶

VI.1. Train-Test-Split for Random Forest model¶

In [93]:
# Splitting the data into train and test sets in 70:30 ratio
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.30, stratify=y )
In [94]:
X_train.shape, X_test.shape
Out[94]:
((31647, 51), (13564, 51))

Over Sampling Using SMOTE

In [95]:
sm = SMOTE(random_state=12)
X_train_r, y_train_r = sm.fit_resample(X_train, y_train)

VI.2. Building random forest model¶

In [96]:
clf_rf = RandomForestClassifier(n_estimators=110, random_state=42)
clf_rf.fit(X_train_r, y_train_r)
Out[96]:
RandomForestClassifier(n_estimators=110, random_state=42)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
RandomForestClassifier(n_estimators=110, random_state=42)

VI.3. Checking random forest model performance on the training set¶

In [97]:
confusion_matrix_sklearn(clf_rf, X_train_r, y_train_r)
In [98]:
rf_perf_train = model_performance_classification_sklearn(
    clf_rf, X_train_r, y_train_r
)
rf_perf_train
Out[98]:
Accuracy Recall Precision F1
0 1.00000 1.00000 1.00000 1.00000

The model is overfitting on train dataset

VI.4. Checking random forest model performance on the testing set¶

In [99]:
confusion_matrix_sklearn(clf_rf, X_test, y_test)
In [100]:
rf_perf_test = model_performance_classification_sklearn(
    clf_rf, X_test, y_test
)
rf_perf_test
Out[100]:
Accuracy Recall Precision F1
0 0.90556 0.43163 0.64380 0.51679

The model has a high accuracy compared to the precision

Plotting the feature importance of each variable

In [101]:
index_best_model = np.argmax(f1_test)
best_model = clfs[index_best_model]
print(best_model)

DecisionTreeClassifier(ccp_alpha=7.258901846204399e-05, class_weight='balanced',
                       random_state=1)
In [102]:
importances = best_model.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12, 12))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="Blue", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

Even in random forest model, duration and poutcome are the most important features

In [103]:
# training performance comparison

models_train_comp_df = pd.concat(
    [
        decision_tree_perf_train.T,
        decision_tree_tune_perf_train.T,
        decision_tree_post_perf_train.T,
        rf_perf_train.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
    "Random Forest (resampled)",
]

# test set  performance comparison

models_test_comp_df = pd.concat(
    [
        decision_tree_perf_test.T,
        decision_tree_tune_perf_test.T,
        decision_tree_post_test.T,
        rf_perf_test.T,

    ],
    axis=1,
)
models_test_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
    "Random Forest (resampled)",
]
In [104]:
models_train_comp_df.T
Out[104]:
Accuracy Recall Precision F1
Decision Tree sklearn 1.00000 1.00000 1.00000 1.00000
Decision Tree (Pre-Pruning) 0.82513 0.82772 0.38753 0.52790
Decision Tree (Post-Pruning) 0.93127 0.99866 0.63239 0.77440
Random Forest (resampled) 1.00000 1.00000 1.00000 1.00000

Distributions

2-d distributions

Values

In [113]:
from matplotlib import pyplot as plt
_df_10['F1'].plot(kind='line', color='#0000FF', figsize=(10, 4), title='F1')
plt.gca().spines[['top', 'right']].set_visible(False)
In [114]:
from matplotlib import pyplot as plt
_df_9['Precision'].plot(kind='line', color='#0000FF',figsize=(10, 4), title='Precision')
plt.gca().spines[['top', 'right']].set_visible(False)
In [110]:
from matplotlib import pyplot as plt
_df_8['Recall'].plot(kind='line', color='#0000FF', figsize=(10, 4), title='Recall')
plt.gca().spines[['top', 'right']].set_visible(False)
In [108]:
from matplotlib import pyplot as plt
_df_7['Accuracy'].plot(kind='line', color='#0000FF', figsize=(10, 4), title='Accuracy')
plt.gca().spines[['top', 'right']].set_visible(False)
# reset the plot configurations to default
plt.rcdefaults()
plt.show()
In [106]:
models_test_comp_df.T
Out[106]:
Accuracy Recall Precision F1
Decision Tree sklearn 0.87651 0.49323 0.46252 0.47738
Decision Tree (Pre-Pruning) 0.82159 0.81560 0.37217 0.51111
Decision Tree (Post-Pruning) 0.86597 0.73050 0.44730 0.55485
Random Forest (resampled) 0.90556 0.43163 0.64380 0.51679

Distributions

2-d distributions

Values

In [123]:
from matplotlib import pyplot as plt
_df_21['F1'].plot(kind='line', figsize=(10, 4), title='F1')
plt.gca().spines[['top', 'right']].set_visible(False)
In [122]:
from matplotlib import pyplot as plt
_df_20['Precision'].plot(kind='line', figsize=(10, 4), title='Precision')
plt.gca().spines[['top', 'right']].set_visible(False)
In [121]:
from matplotlib import pyplot as plt
_df_19['Recall'].plot(kind='line', figsize=(10, 4), title='Recall')
plt.gca().spines[['top', 'right']].set_visible(False)
In [124]:
from matplotlib import pyplot as plt
_df_18['Accuracy'].plot(kind='line', figsize=(10, 4), title='Accuracy')
plt.gca().spines[['top', 'right']].set_visible(False)

Random forest has a high accuracy compared to Decision Tree sklearn model, using the test dataset

MODEL EVALUATION AND CONCLUSION¶

Looking back at the Exploration Data Analysis, Data are highly imbalanced Duration remains the most important feature. We also think that candidates that was contacted during the previous campaign, were most likely to subscribe or deposit after being contacted during the forthcoming campaigns.

Married participated the most in the campaign, followed by single then finally divorced. This clearly illustrates that married people care most about savings, deposit etc compared to the others

In term of job categories, unknown, technicians, managers, self-employed, maid participated mostly in the campaign

Students and retired have not really participated compared to the others, may be because they likely don’t have relative and income.

Random forest (resampled) seems to be the better model with high accuracy on test dataset. We think its better performance may be due to the fact that data were balanced using SMOTE (Synthetic Minority Oversampling TEchnique)